Senior/Staff ML Engineer, ML Acceleration and Performance

About Rivian Rivian is on a mission to keep the world adventurous forever. This goes for the emissions-free Electric Adventure Vehicles we build, and the curious, courageous souls we seek to attract. As a company, we constantly challenge what’s possible, never simply accepting what has always been done. We reframe old problems, seek new solutions and operate comfortably in areas that are unknown. Our backgrounds are diverse, but our team shares a love of the outdoors and a desire to protect it for future generations.

Summary As a Staff Software Engineer for ML Optimization and Hardware Acceleration, you will be a lead member of the Autonomy team at Rivian. You will develop and optimize advanced machine learning algorithms that directly impact the safety-critical self-driving features of our category-defining vehicles. This role focuses on the intersection of cutting-edge model architectures including Transformers, LLMs, VLMs, LDMs and high-performance hardware execution. You will bridge the gap between theoretical ML research and real- time embedded deployment, ensuring our autonomy stack remains both state-of-the-art and ultra-efficient.

Model Optimization: Develop and deploy ultra-low latency Deep Learning and Machine Learning algorithms specifically tailored for Rivian ADAS and Autonomy use cases. Hardware-Aware Design: Research and implement hardware-aware optimization strategies, including Post-Training Quantization (PTQ), Quantization-Aware Training (QAT), kernel fusion, and model distillation to maximize throughput on embedded platforms. Performance Profiling: Utilize and automate deep-dive profiling tools (e.g., Torch Profile, NVIDIA Nsight) to identify bottlenecks and ensure performance consistency across weekly evaluation runs. Cross-Functional Collaboration: Partner with low-level software and hardware architecture teams to characterize in-house ML models on embedded platforms, optimizing them within strict compute and memory constraints. Architectural Reasoning: Apply a deep understanding of GPU architectures to optimize models across significantly different hardware targets, ensuring scalability across the Rivian fleet. Workflow and Infrastructure Engineering: Design and build automated pipelines for regular model profiling across diverse architectures to enhance organization-wide insight into execution bottlenecks.

Education/Experience: MS (+3 years of experience in deep learning, heterogeneous computing, and ML accelerators) or Ph.D. in Computer Science, Electrical Engineering, or a related field. Core ML Expertise: Deep understanding of modern model architectures, including Transformers, LLMs, VLMs and LDMs. Optimization Skills: Proven experience in model compression techniques: knowledge distillation, pruning, and quantization (PTQ/QAT). Hardware Knowledge: In-depth understanding of GPU architecture and the ability to optimize for diverse hardware specifications. Technical Toolset: ○ Proficiency in Python and deep knowledge of PyTorch or TensorFlow. ○ Hands-on experience with TensorRT, AIMET, ONNX runtimes. ○ Experience with low-level programming (CUDA kernels, C++, or BLAS subroutines) for inference logic. ○ Experience with profiling tools like torch profiler and nvidia nsight. Leadership: Strong team player with excellent communication skills to drive complex, cross-functional efforts in a fast-paced environment. How to distinguish yourself: ○ A strong track record of publications in top-tier venues such as MLSys, ICML, NeurIPS, or ISCA. ○ Significant and direct industry experience in a related domain. ○ Active participation and contributions to relevant open-source projects. ○ Public demonstrations of expertise, including technical talks, presentations, or live demos. Pay Disclosure Salary Range for California Based Applicants: $228,000 - $285,000 (actual compensation will be determined based on experience, location, and other factors permitted by